Obesity increases the risk of developing cancer, particularly hormone-dependent cancer. Fifty-four percent of Americans are overweight or obese; women and ethnic minorities have the highest rates. While obesity is the result of an imbalance between energy intake and energy expenditure, the etiology of obesity differs from person to person. Thus, no single obesity prevention or treatment strategy will help every person; tailored interventions hold the greatest promise. We therefore propose to link genetic, hormonal, and behavioral factors into individual obesity risk profiles to identify those at risk of developing obesity and the best interventions for them. We will capitalize on a unique data resource, the previously collected demographic, dietary, and behavioral data and plasma and DNA samples from 550 European Americans and 225 African Americans participating in the Women's Health Initiative trial. Our first aim is to ascertain the relationships between body mass index (BMI; =weight in kg /height in m2) and (i) putative genetic determinants of obesity-related behavior, specifically two restriction fragment length polymorphisms of the D2 dopamine receptor gene (DRD2); (ii) hormonal determinants of obesity, specifically leptin, insulin-like growth factor 1 (IGF-1) and its major binding protein (BP-3), dehydroepiandrosterone (DHEA) and its sulfated conjugate (DHEAS), estrone, estradiol, and sex hormone-binding globulin (SHBG); and (iii) behavioral determinants of obesity, including tobacco use; excessive alcohol use; high dietary intake of total calories, total fat, and total carbohydrates; low levels of physical activity; and depression. We will also construct a comprehensive obesity risk assessment profile of the most important genetic, hormonal, and behavioral determinants we identify.